基于Keras实现Kaggle2013--Dogs vs. Cats12500张猫狗图像的精准分类

【下载数据集】

• 下载链接–百度网盘
  

【整理数据集】

• 将训练数据集分割成训练集、验证集、测试集，目录结构如图所示：
  

• 在Pycharm中新建项目，创建split_dataset.py

import os, shutil

# 数据集解压之后的目录
original_dataset_dir = 'D:\kaggle\dogsvscats\\train'
# 存放小数据集的目录
base_dir = 'D:\kaggle\dogsvscats\\cats_and_dogs_small'
os.mkdir(base_dir)

# 建立训练集、验证集、测试集目录
train_dir = os.path.join(base_dir, 'train')
os.mkdir(train_dir)
validation_dir = os.path.join(base_dir, 'validation')
os.mkdir(validation_dir)
test_dir = os.path.join(base_dir, 'test')
os.mkdir(test_dir)

# 将猫狗照片按照训练、验证、测试分类
train_cats_dir = os.path.join(train_dir, 'cats')
os.mkdir(train_cats_dir)

train_dogs_dir = os.path.join(train_dir, 'dogs')
os.mkdir(train_dogs_dir)

validation_cats_dir = os.path.join(validation_dir, 'cats')
os.mkdir(validation_cats_dir)

validation_dogs_dir = os.path.join(validation_dir, 'dogs')
os.mkdir(validation_dogs_dir)

test_cats_dir = os.path.join(test_dir, 'cats')
os.mkdir(test_cats_dir)

test_dogs_dir = os.path.join(test_dir, 'dogs')
os.mkdir(test_dogs_dir)

# 切割数据集
fnames = ['cat.{}.jpg'.format(i) for i in range(1000)]
for fname in fnames:
    src = os.path.join(original_dataset_dir, fname)
    dat = os.path.join(train_cats_dir, fname)
    shutil.copyfile(src, dat)

fnames = ['cat.{}.jpg'.format(i) for i in range(1000, 1500)]
for fname in fnames:
    src = os.path.join(original_dataset_dir, fname)
    dat = os.path.join(validation_cats_dir, fname)
    shutil.copyfile(src, dat)

fnames = ['cat.{}.jpg'.format(i) for i in range(1500, 2000)]
for fname in fnames:
    src = os.path.join(original_dataset_dir, fname)
    dat = os.path.join(test_cats_dir, fname)
    shutil.copyfile(src, dat)

fnames = ['dog.{}.jpg'.format(i) for i in range(1000)]
for fname in fnames:
    src = os.path.join(original_dataset_dir, fname)
    dat = os.path.join(train_dogs_dir, fname)
    shutil.copyfile(src, dat)

fnames = ['dog.{}.jpg'.format(i) for i in range(1000, 1500)]
for fname in fnames:
    src = os.path.join(original_dataset_dir, fname)
    dat = os.path.join(validation_dogs_dir, fname)
    shutil.copyfile(src, dat)

fnames = ['dog.{}.jpg'.format(i) for i in range(1500, 2000)]
for fname in fnames:
    src = os.path.join(original_dataset_dir, fname)
    dat = os.path.join(test_dogs_dir, fname)
    shutil.copyfile(src, dat)

【建立简单版CNN网络模型】

from keras import layers
from keras import models
model = models.Sequential()
model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(150, 150, 3)))
model.add(layers.MaxPool2D((2, 2)))

model.add(layers.Conv2D(64, (3, 3), activation='relu'))
model.add(layers.MaxPool2D((2, 2)))

model.add(layers.Conv2D(128, (3, 3), activation='relu'))
model.add(layers.MaxPool2D((2, 2)))

model.add(layers.Conv2D(128, (3, 3), activation='relu'))
model.add(layers.MaxPool2D((2, 2)))

model.add(layers.Flatten())
model.add(layers.Dense(512, activation='relu'))
model.add(layers.Dense(1, activation='sigmoid'))

model.compile(loss='binary_crossentropy', optimizer=optimizers.RMSprop(lr=1e-4), metrics=['acc'])

【对图像信息进行预处理】

• 读取图片文件；
• 将jpg解码成RGB像素点；
• 将这些像素点转换成浮点型张量；
• 将[0, 255]区间的像素值减小到[0, 1]区间中，CNN更喜欢处理小的输入值。

train_datagen = ImageDataGenerator(rescale=1./255)
test_datagen = ImageDataGenerator(rescale=1./255)

train_generator = train_datagen.flow_from_directory(
    directory=train_dir,
    target_size=(150, 150),
    batch_size=20,
    class_mode='binary')

validation_generator = test_datagen.flow_from_directory(
    directory=validation_dir,
    target_size=(150, 150),
    batch_size=20,
    class_mode='binary')

• 用fit_generator向模型中填充数据

history = model.fit_generator(
    train_generator,
    steps_per_epoch=100,
    epochs=30,
    validation_data=validation_generator,
    validation_steps=50)

• 保存模型

model.save('cats_and_dogs_small_1.h5')

• 显示训练中loss和acc的曲线

acc = history.history['acc']
val_acc = history.history['val_acc']
loss = history.history['loss']
val_loss = history.history['val_loss']

epochs = range(1, len(acc) + 1)
plt.plot(epochs, acc, 'bo', label='Training acc')
plt.plot(epochs, val_acc, 'b', label='Validation acc')
plt.title('Training and validation accuracy')
plt.legend()

plt.figure()

plt.plot(epochs, loss, 'bo', label='Training loss')
plt.plot(epochs, val_loss, 'b', label='Validation loss')
plt.title('Training and validation loss')
plt.legend()

plt.show()

【简单版CNN模型完整代码】

from keras import layers
from